The virtual microbiome: A computational framework to evaluate microbiome analyses

Abstract

Microbiomes have been the focus of a substantial research effort in the last decades. The composition of microbial populations is normally determined by comparing DNA sequences sampled from those populations with the sequences stored in genomic databases. Therefore, the amount of information available in databanks should be expected to constrain the accuracy of microbiome analyses. Albeit normally ignored in microbiome studies, this constraint could severely compromise the reliability of microbiome data. To test this hypothesis, we generated virtual bacterial populations that exhibit the ecological structure of real-world microbiomes. Confronting the analyses of virtual microbiomes with their original composition revealed critical issues in the current approach to characterizing microbiomes, issues that were empirically confirmed by analyzing the microbiome of Galleria mellonella larvae. To reduce the uncertainty of microbiome data, the effort in the field must be channeled towards significantly increasing the amount of available genomic information and optimizing the use of this information.

Long amplicon nanopore sequencing of Botrytis cinerea and other fungal species present in infected grapevine leaf samples

“…Further field experiments are needed with mock communities to establish whether the method is providing a real abundance of species. To exclude false-positive results, the enrichment of databases with closely related species sequences within different genera is crucial [ 21 ].…”

Plastic degradation by insect hexamerins: Near-atomic resolution structures of the polyethylene-degrading proteins from the wax worm saliva

“…Despite the intense effort in this direction, the data collected so far resulted inconclusive, and doubts have been shed on the primary role of microorganisms in the insect larvae’s ability ( 13 , 23 ). In this line, recent studies further questioned a primary role of the invertebrates’ microbiome ( 24 ), showing that the larvae of G. mellonella oxidize PE using their own saliva, defined as the liquid collected from the buccal opening ( 21 ). In particular, proteins of the hexamerin family present in the insect saliva proved to oxidize PE.…”